Regression Model using Ranger for FeNi

1. Reading the dataset

library(readr)
dataset <- read_csv("new_dataset.csv")
head(dataset)

2. Splitting the dataset

We will split the dataset into a training set (70%) and a testing set (30%).

set.seed(123) # for reproducibility
sample_index <- sample(1:nrow(dataset), 0.7*nrow(dataset))
train_data <- dataset[sample_index, ]
test_data <- dataset[-sample_index, ]

3. Creating a regression model

Using the ranger package, we’ll predict the tmg feature and using permutation

library(ranger)
model_per <- ranger(tmg ~ ., data = train_data, importance = 'permutation')
model_per

4. Calculate importance with permutation

library(dplyr)
library(ggplot2)

plot_perm <- importance(model_per) |> as.data.frame() |> add_rownames("predictor") |> mutate(importance=`importance(model_per)`) |> select(predictor,importance)|> arrange(desc(importance)) |> head(20) |> mutate(predictor = factor(predictor, levels = rev(unique(predictor))))  |>
  ggplot()+
  geom_col(aes(y=predictor,x=importance),fill='darkblue', color='gray')+
  ggtitle("Top 20 predictor importance using permutation")+
  theme_minimal()
plot_perm

5. calculate importante with impurity

library(ranger)
model_imp <- ranger(tmg ~ ., data = train_data, importance = 'impurity')
model_imp

plot_imp<-importance(model_imp) |> as.data.frame() |> add_rownames("predictor") |> mutate(importance=`importance(model_imp)`) |> select(predictor,importance)|> arrange(desc(importance)) |> head(20) |> mutate(predictor = factor(predictor, levels = rev(unique(predictor))))  |>
  ggplot()+
  geom_col(aes(y=predictor,x=importance),fill='darkblue', color='gray')+
  ggtitle("Top 20 predictor importance using impurity")+
  theme_minimal()

Permutation vs. Impurity

library(gridExtra)
gridExtra::grid.arrange(plot_imp,plot_perm, ncol=2)
plot_imp
plot_perm

5. Evaluate results on test dataset

predictions <- predict(model_per, data = test_data)$predictions
# Compute the RMSE (Root Mean Square Error)
RMSE <- sqrt(mean((predictions - test_data$tmg)^2))
RMSE
predictions <- predict(model_per, data = dataset)$predictions
# Compute the RMSE (Root Mean Square Error)
RMSE <- sqrt(mean((predictions - dataset$tmg)^2))
RMSE

6. Plot: Predicted vs Reference values

library(ggplot2)

predictions <- predict(model_per, data = test_data)$predictions
results <- data.frame(Reference = test_data$tmg, Predicted = predictions)
ggplot(results, aes(x = Reference, y = Predicted)) +
  geom_point(color='blue') +
  #geom_smooth(method = 'lm', color = 'red') +
  geom_abline(intercept = 0,slope =1,color='red')+
  ggtitle("Predicted vs Reference values") +
  #ylim(0,1)+
  xlab("Reference Values") +
  ylab("Predicted Values")+
  theme_bw()

predictions <- predict(model_per, data = train_data)$predictions
results <- data.frame(Reference = train_data$tmg, Predicted = predictions)
ggplot(results, aes(x = Reference, y = Predicted)) +
  geom_point(color='blue') +
  #geom_smooth(method = 'lm', color = 'red') +
  geom_abline(intercept = 0,slope =1,color='red')+
  ggtitle("Predicted vs Reference values") +
  #ylim(0,1)+
  xlab("Reference Values") +
  ylab("Predicted Values")+
  theme_bw()

predictions <- predict(model_per, data = dataset)$predictions
results <- data.frame(Reference = dataset$tmg, Predicted = predictions)
ggplot(results, aes(x = Reference, y = Predicted)) +
  geom_point(color='blue') +
  #geom_smooth(method = 'lm', color = 'red') +
  geom_abline(intercept = 0,slope =1,color='red')+
  ggtitle("Predicted vs Reference values") +
  #ylim(0,1)+
  xlab("Reference Values") +
  ylab("Predicted Values")+
  theme_bw()

# Install and load necessary packages
# install.packages("ranger")
# install.packages("caret")
library(ranger)
library(caret)
library(readr)
library(caret)
library(ranger)
library(caret)
library(dplyr)

library(readr)
library(caret)

Loading required package: ggplot2
Loading required package: lattice

library(ranger)
library(caret)
library(dplyr)

Attaching package: ‘dplyr’

The following objects are masked from ‘package:stats’:

    filter, lag

The following objects are masked from ‘package:base’:

    intersect, setdiff, setequal, union

dataset <- read_csv("new_dataset.csv")

Rows: 568 Columns: 648── Column specification ─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: ","
chr   (1): name
dbl (647): psd_1, psd_2, psd_3, psd_4, psd_5, psd_6, psd_7, psd_8, psd_9, psd_10, psd_11, psd_12, psd_13, psd_14, psd_15, psd_16, psd_17, psd_18, psd_19, psd_20, psd_21, psd_22, psd_23,...
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

nzv <- nearZeroVar(dataset, saveMetrics = FALSE)
dataset <- dataset[, -nzv]
dataset$name <- NULL
set.seed(123) # for reproducibility
sample_index <- sample(1:nrow(dataset), 0.7*nrow(dataset))
train_data <- dataset[sample_index, ]
test_data <- dataset[-sample_index, ]
# mod_2=rgcv(tmg ~ .,data = train_data, importance = 'permutation', cv.fold=10)
model_per <- train(
  tmg ~ .,
  data = train_data,
  method = "ranger",
  trControl = trainControl(method = "cv", number = 10),
  importance = "permutation"
)
# perm_imp <- permutationFeature(model_per, metric = "RMSE", trControl = ctrl)

# model_per <- ranger(tmg ~ .,data = train_data, importance = 'permutation')

do_plot <- function(data, type) {
  predictions <- predict(model_per, data)
  print(data)
  print(predictions)
  print(data$tmg)
  RMSE <- sqrt(mean((predictions - data$tmg)^2))
  results <- data.frame(Reference = data$tmg, Predicted = as.vector(predictions))
  return(ggplot(results, aes(x = Reference, y = Predicted)) +
    geom_point(color='blue') +
    geom_abline(intercept = 0, slope = 1, color='red') +
    ggtitle(paste(type, "Random Forest; RMSE:", RMSE)) +
    xlab("Reference Values") +
    ylab("Predicted Values") +
    theme_bw())
}
do_plot(test_data, "Test")

  [1] 0.2689269 0.4053836 0.3649996 0.3667678 0.3655683 0.3632867 0.3644209 0.3538282 0.3481348 0.3520358 0.2509988 0.2582524 0.2423742 0.2702699 0.2320738 0.2624782 0.2573334 0.2534757
 [19] 0.4355293 0.4039390 0.4052275 0.3981909 0.4023658 0.4070284 0.4013852 0.4037824 0.4374316 0.4402415 0.4429377 0.4425301 0.4406133 0.4429935 0.4418933 0.4424070 0.3936540 0.4019863
 [37] 0.4031244 0.3733902 0.3283978 0.3555004 0.3969559 0.3450029 0.3259292 0.3420153 0.3383234 0.3451772 0.3478461 0.3526765 0.3430144 0.3446427 0.3341462 0.3891644 0.4637248 0.3492502
 [55] 0.3741866 0.3745212 0.3693188 0.3760095 0.3670509 0.3733676 0.3731008 0.3797560 0.4218498 0.4209190 0.4210648 0.4214584 0.3809959 0.3766756 0.3782384 0.4339579 0.2823448 0.2630980
 [73] 0.2755730 0.2577974 0.4198652 0.4244795 0.4206386 0.4262408 0.4218937 0.4178009 0.3916835 0.3930787 0.3927441 0.3925250 0.3875441 0.3979316 0.3691211 0.3988731 0.3948259 0.3590153
 [91] 0.3595331 0.3533867 0.4846511 0.2310049 0.2133585 0.2255054 0.4201381 0.4239103 0.4178412 0.4110824 0.3562412 0.3442412 0.3515870 0.3597254 0.3563275 0.3526094 0.3524258 0.3509225
[109] 0.4186149 0.4196975 0.4220581 0.4237472 0.4237661 0.4246864 0.4252443 0.4259567 0.2363466 0.2322584 0.2338668 0.2313073 0.2574623 0.2548630 0.2602922 0.2597568 0.2266073 0.2271369
[127] 0.2221563 0.2232829 0.2573293 0.2662636 0.2580088 0.2470324 0.4281665 0.4082926 0.4060756 0.4070808 0.4083692 0.3539352 0.3563792 0.3563985 0.3574177 0.3612098 0.3551919 0.4071818
[145] 0.4031748 0.4107778 0.4057403 0.2605003 0.2599739 0.2659934 0.2595775 0.2646526 0.2620849 0.2688872 0.2634987 0.2582253 0.4829680 0.4506270 0.4503725 0.2692258 0.2754180 0.4301141
[163] 0.4779121 0.4761561 0.4372146 0.3746027 0.4324379 0.3444640 0.2238289 0.4754487 0.4089247
  [1] 0.2100 0.4400 0.3700 0.3800 0.3800 0.3800 0.3700 0.3600 0.3500 0.3400 0.2800 0.2700 0.2300 0.2400 0.2200 0.2500 0.2100 0.2400 0.4800 0.4100 0.4000 0.4000 0.4100 0.4000 0.4100 0.4100
 [27] 0.4500 0.4400 0.4400 0.4400 0.4500 0.4400 0.4500 0.4500 0.4100 0.4200 0.4000 0.4300 0.3700 0.4200 0.3000 0.3400 0.3400 0.3600 0.3300 0.3600 0.3400 0.3500 0.3300 0.3200 0.3200 0.3000
 [53] 0.4600 0.3800 0.3800 0.3700 0.3900 0.3900 0.3700 0.3500 0.3800 0.3800 0.4300 0.4300 0.4300 0.4300 0.4000 0.3800 0.3900 0.4900 0.3200 0.2500 0.2900 0.2400 0.4300 0.4300 0.4300 0.4300
 [79] 0.4300 0.4300 0.3800 0.3800 0.3900 0.3900 0.3800 0.4300 0.3200 0.3800 0.2900 0.4300 0.3600 0.3900 0.5000 0.1900 0.2800 0.1300 0.4100 0.4300 0.4100 0.4200 0.3800 0.3600 0.3800 0.3800
[105] 0.3500 0.3500 0.3700 0.3600 0.4300 0.4000 0.4100 0.4200 0.4000 0.4300 0.4000 0.4100 0.2200 0.2700 0.2000 0.2400 0.2700 0.2400 0.3100 0.1600 0.2800 0.2600 0.2000 0.2300 0.2600 0.3000
[131] 0.2800 0.2200 0.4400 0.4000 0.4100 0.4000 0.4000 0.3500 0.3500 0.3600 0.3700 0.3600 0.3800 0.4000 0.4100 0.3900 0.4000 0.3000 0.1500 0.3100 0.2600 0.2700 0.2100 0.3100 0.2200 0.2900
[157] 0.5266 0.4806 0.4900 0.1926 0.1979 0.4837 0.4868 0.4965 0.4148 0.4023 0.3370 0.3539 0.2480 0.5105 0.4037

do_plot(train_data, "Train")

  [1] 0.2871247 0.4081337 0.3826972 0.2377298 0.4016557 0.4357830 0.3524646 0.2504880 0.4615884 0.3672767 0.3558274 0.4458457 0.4255670 0.3530934 0.3368018 0.2797827 0.2251930 0.4264190
 [19] 0.3419663 0.3470506 0.3523341 0.3646444 0.3503391 0.3699822 0.2610450 0.3616606 0.3503485 0.3865345 0.3659097 0.4078245 0.3474856 0.3358104 0.3968721 0.3804572 0.4483053 0.2753050
 [37] 0.2112019 0.4413384 0.4517705 0.3783560 0.2668398 0.4078221 0.4685421 0.2016838 0.3293283 0.4142457 0.4318029 0.3544025 0.4079328 0.3690654 0.4122699 0.2465563 0.3673308 0.5016549
 [55] 0.4451141 0.3887534 0.2176683 0.3487803 0.3672408 0.1910486 0.2789166 0.3493443 0.3958754 0.3514727 0.3980844 0.3641106 0.2369651 0.3649318 0.3580202 0.4193415 0.4015868 0.3560847
 [73] 0.3659287 0.4234297 0.4394534 0.3482001 0.3766881 0.2478012 0.4446557 0.3562384 0.3419427 0.2818055 0.4456081 0.1912018 0.3750578 0.4326722 0.4045210 0.3903895 0.3610422 0.3893311
 [91] 0.2709527 0.3585730 0.3672459 0.2696114 0.3181830 0.3504564 0.2313074 0.3611232 0.3518126 0.5090696 0.4078832 0.4057196 0.4297858 0.3943910 0.3414150 0.3410438 0.4413633 0.4216207
[109] 0.3525913 0.4030051 0.3516369 0.4878344 0.3725990 0.2618794 0.3958660 0.4034735 0.4057248 0.4309828 0.4846070 0.2118861 0.2577723 0.2793681 0.4292147 0.4168836 0.2714244 0.3573873
[127] 0.3971263 0.3870580 0.2655985 0.4377528 0.4390349 0.3802004 0.3566058 0.2488558 0.4569518 0.4303637 0.3816910 0.4010004 0.4399207 0.2634908 0.4194226 0.2432385 0.3426440 0.3603614
[145] 0.4731791 0.4267103 0.3615027 0.1961402 0.3581327 0.3470920 0.4237846 0.2102208 0.4937867 0.4464279 0.3725019 0.4142353 0.4326623 0.3525644 0.3709085 0.3448214 0.4992390 0.4036429
[163] 0.2631131 0.4272854 0.4281141 0.4317613 0.3573665 0.4226035 0.3623376 0.3640332 0.3612580 0.4315841 0.3495268 0.2184150 0.3666888 0.2225038 0.4354059 0.3977908 0.3986227 0.3811427
[181] 0.4076734 0.3451853 0.3324869 0.2236381 0.4838027 0.4524149 0.2353579 0.3888018 0.3707591 0.4267259 0.4099445 0.4291577 0.3769071 0.4272276 0.4195955 0.3676444 0.3921727 0.3477877
[199] 0.2594514 0.4782895 0.3697770 0.4149692 0.2393905 0.4825637 0.4040740 0.3678768 0.5012013 0.3915276 0.3248728 0.4251447 0.3999874 0.3786974 0.4101786 0.4101221 0.4259970 0.4231428
[217] 0.4142954 0.3460745 0.3738009 0.4600988 0.4094892 0.4082986 0.3532721 0.3741668 0.4020766 0.3531541 0.4144587 0.3303721 0.3375340 0.2426554 0.3279041 0.3996086 0.3986557 0.3523415
[235] 0.3767642 0.3988434 0.3450894 0.3583992 0.2748303 0.3172573 0.2219365 0.2107871 0.3433137 0.3766790 0.2275663 0.3661834 0.4652893 0.2509934 0.3524690 0.4014159 0.1961742 0.1852804
[253] 0.4264948 0.4411043 0.4274671 0.2582569 0.4689274 0.3488020 0.3941227 0.3555659 0.3772977 0.4331665 0.3908616 0.1926758 0.2416522 0.2739350 0.3967536 0.4410547 0.2200760 0.4011318
[271] 0.3460387 0.2298082 0.3959045 0.3701122 0.4205830 0.2841555 0.4016048 0.4399034 0.2515936 0.3997106 0.3959717 0.4463507 0.2559964 0.1962145 0.2473089 0.2087417 0.2226538 0.4111853
[289] 0.2430133 0.2224785 0.2811484 0.3713830 0.2411744 0.3450342 0.2366318 0.4662495 0.3516145 0.4419756 0.3692998 0.3845447 0.3712363 0.2479043 0.4188207 0.4205333 0.4425131 0.3598113
[307] 0.4148218 0.3678192 0.2434608 0.4134862 0.3470594 0.3547905 0.3799720 0.2169570 0.3540527 0.3927499 0.4089690 0.2038637 0.4203946 0.2572152 0.3523329 0.4128596 0.4260638 0.3390340
[325] 0.4936290 0.3571270 0.3647404 0.2095205 0.3668304 0.3011025 0.4184188 0.3355538 0.4356104 0.4635185 0.2182548 0.4591063 0.4123895 0.3860391 0.2595910 0.3555031 0.4019427 0.3466801
[343] 0.4414926 0.5060413 0.3988787 0.4411182 0.4367580 0.4036836 0.2205215 0.3497740 0.3413632 0.3604704 0.4376736 0.4650387 0.2086093 0.2118318 0.3822077 0.2392990 0.3423928 0.3424705
[361] 0.3877463 0.3696465 0.4010527 0.4765927 0.3379579 0.2859952 0.3931068 0.4765927 0.3919439 0.1923227 0.4263084 0.3748105 0.3681300 0.2516127 0.3678701 0.4091938 0.2414272 0.4389744
[379] 0.2600344 0.3529609 0.4161375 0.2195732 0.3509037 0.3603447 0.4154150 0.3838294 0.4111321 0.3609444 0.2735856 0.3790042 0.4046832 0.1955575 0.2544728 0.2671579 0.4679386 0.3983077
[397] 0.4337486
  [1] 0.3100 0.4100 0.3900 0.2200 0.4500 0.4400 0.3500 0.2400 0.4600 0.3700 0.3600 0.4500 0.4300 0.3500 0.3300 0.2900 0.2200 0.4300 0.3300 0.3500 0.3623 0.3700 0.3500 0.3800 0.2000 0.3500
 [27] 0.3500 0.3900 0.3800 0.4100 0.3500 0.3300 0.4100 0.4200 0.4485 0.3100 0.2000 0.4400 0.4700 0.3800 0.2700 0.4100 0.4900 0.1700 0.3200 0.4100 0.4305 0.3600 0.4100 0.3700 0.4200 0.2600
 [53] 0.3700 0.5000 0.4500 0.3700 0.1900 0.3500 0.4200 0.1600 0.3000 0.3200 0.3900 0.3600 0.4000 0.3700 0.2400 0.3700 0.3600 0.4200 0.4000 0.3700 0.3700 0.4300 0.4400 0.3500 0.3500 0.2200
 [79] 0.4500 0.3100 0.3300 0.3000 0.4500 0.1700 0.3800 0.4400 0.4000 0.3900 0.3400 0.3900 0.2700 0.3600 0.3700 0.2800 0.3200 0.3500 0.2100 0.3700 0.3500 0.5200 0.4100 0.4100 0.4300 0.4247
[105] 0.3300 0.3400 0.4400 0.4200 0.3600 0.4000 0.3500 0.5045 0.3800 0.2800 0.4000 0.4000 0.4200 0.4438 0.4900 0.2100 0.2600 0.2800 0.4300 0.4100 0.2800 0.3600 0.3900 0.4000 0.2800 0.4400
[131] 0.4400 0.3800 0.3600 0.2500 0.4700 0.4300 0.4000 0.4000 0.4400 0.2800 0.4200 0.2400 0.3400 0.3600 0.4794 0.4300 0.3700 0.1300 0.3600 0.3400 0.4700 0.1900 0.5000 0.4500 0.3800 0.4200
[157] 0.4400 0.3500 0.3800 0.3400 0.5400 0.4000 0.2900 0.4300 0.4300 0.4400 0.3600 0.4200 0.3600 0.3700 0.3600 0.4400 0.3500 0.2100 0.3700 0.2100 0.4400 0.4000 0.4000 0.3900 0.4100 0.3400
[183] 0.3200 0.2300 0.4900 0.4546 0.2200 0.3700 0.3700 0.4300 0.4200 0.4300 0.3768 0.4300 0.4200 0.3800 0.3900 0.3500 0.2700 0.4900 0.3700 0.4200 0.2500 0.4800 0.4000 0.3700 0.5100 0.3900
[209] 0.2800 0.4300 0.4200 0.3800 0.4100 0.4000 0.4300 0.4400 0.4100 0.3400 0.3800 0.4600 0.4100 0.4249 0.3500 0.3700 0.4000 0.3600 0.4200 0.3200 0.3300 0.2400 0.3200 0.4000 0.4100 0.3500
[235] 0.3800 0.4000 0.3400 0.3600 0.2900 0.2993 0.1900 0.1900 0.3400 0.3500 0.2300 0.3700 0.4700 0.2600 0.3600 0.4000 0.1700 0.1600 0.4300 0.4400 0.4300 0.2600 0.5036 0.3500 0.4100 0.3500
[261] 0.3800 0.4300 0.3900 0.1700 0.2400 0.2800 0.3900 0.4400 0.2200 0.4000 0.3500 0.2200 0.3900 0.3800 0.4200 0.1800 0.4309 0.4400 0.2500 0.4000 0.4100 0.4500 0.2600 0.1700 0.2400 0.1800
[287] 0.2200 0.4200 0.2500 0.2000 0.2900 0.3800 0.2500 0.3400 0.2600 0.4700 0.3500 0.4400 0.3800 0.3800 0.3800 0.2400 0.4200 0.4200 0.4500 0.3600 0.4200 0.3700 0.2400 0.4200 0.3400 0.3600
[313] 0.3600 0.2000 0.3500 0.3900 0.4100 0.1900 0.4200 0.2600 0.3500 0.4100 0.4300 0.3400 0.5000 0.3500 0.3600 0.2100 0.3700 0.3000 0.4200 0.3200 0.4548 0.4600 0.2200 0.4600 0.4200 0.3900
[339] 0.2355 0.3600 0.4000 0.3800 0.4400 0.5100 0.4000 0.4400 0.4400 0.4000 0.2300 0.3500 0.3300 0.3600 0.4400 0.4700 0.1900 0.1900 0.4000 0.2300 0.3300 0.3400 0.3900 0.3700 0.4000 0.4800
[365] 0.3300 0.2900 0.3900 0.4800 0.3900 0.1717 0.4300 0.3600 0.3700 0.2700 0.3700 0.4100 0.2500 0.4400 0.2600 0.3500 0.4200 0.2000 0.3400 0.3600 0.4200 0.3900 0.4200 0.3600 0.2758 0.3800
[391] 0.4163 0.1456 0.2700 0.2800 0.4800 0.4200 0.4400

do_plot(dataset, "Total")

  [1] 0.2689269 0.4053836 0.3466801 0.3986557 0.3725990 0.3681300 0.3738009 0.3696465 0.3649996 0.3666888 0.3725019 0.3571270 0.3690654 0.3672767 0.3667678 0.3612580 0.3655683 0.3632867
 [19] 0.3644209 0.3426440 0.3538282 0.3470920 0.3503391 0.3488020 0.3477877 0.3368018 0.3460745 0.3481348 0.3379579 0.3450342 0.3547905 0.3520358 0.3495268 0.3544025 0.2509988 0.2200760
 [37] 0.2582524 0.2423742 0.2016838 0.2184150 0.2753050 0.2102208 0.2702699 0.2320738 0.2107871 0.2624782 0.2573334 0.2655985 0.2534757 0.3011025 0.2478012 0.1961402 0.3423928 0.4838027
 [55] 0.4846070 0.4355293 0.4825637 0.4039390 0.4019427 0.4052275 0.4076734 0.3981909 0.4023658 0.3980844 0.4070284 0.3996086 0.4015868 0.4013852 0.4122699 0.3931068 0.4037824 0.4078245
 [73] 0.4399034 0.4446557 0.4411182 0.4331665 0.4377528 0.4374316 0.4402415 0.4410547 0.4425131 0.4429377 0.4390349 0.4464279 0.4394534 0.4425301 0.4406133 0.4389744 0.4451141 0.4413384
 [91] 0.4458457 0.4376736 0.4463507 0.4399207 0.4429935 0.4419756 0.4414926 0.4456081 0.4418933 0.4367580 0.4424070 0.4413633 0.3936540 0.4014159 0.4040740 0.4101786 0.4057196 0.4036836
[109] 0.4019863 0.3958754 0.3959045 0.4031244 0.4078832 0.4010527 0.4099445 0.4078221 0.3988434 0.4357830 0.3748105 0.4111321 0.3493443 0.3860391 0.3838294 0.3733902 0.4354059 0.3283978
[127] 0.4326722 0.4123895 0.3999874 0.3941227 0.3555004 0.3983077 0.3969559 0.3248728 0.2610450 0.3766881 0.3971263 0.3450029 0.3424705 0.3259292 0.3474856 0.3420153 0.3470506 0.3383234
[145] 0.3531541 0.3433137 0.3555031 0.3303721 0.3451772 0.3478461 0.3560847 0.3324869 0.3358104 0.3410438 0.3516369 0.3390340 0.3526765 0.3514727 0.3293283 0.3525913 0.3279041 0.3430144
[163] 0.3375340 0.3615027 0.3482001 0.3503485 0.3446427 0.3816910 0.3341462 0.3766790 0.3891644 0.3959717 0.4637248 0.3610422 0.4662495 0.4652893 0.4237846 0.3562384 0.3826972 0.3492502
[181] 0.3647404 0.3741866 0.3745212 0.3802004 0.3707591 0.3811427 0.3693188 0.3760095 0.3670509 0.3733676 0.3772977 0.3731008 0.3797560 0.4992390 0.4016557 0.4569518 0.4203946 0.4272854
[199] 0.4193415 0.4317613 0.4218498 0.4205333 0.4209190 0.4184188 0.4195955 0.4210648 0.4214584 0.4260638 0.4142457 0.4264948 0.4264190 0.3903895 0.3809959 0.3750578 0.3697770 0.3766756
[217] 0.3865345 0.3786974 0.3790042 0.3782384 0.3877463 0.3741668 0.3783560 0.3616606 0.3767642 0.3870580 0.4339579 0.3822077 0.2504880 0.2479043 0.2823448 0.3181830 0.2630980 0.2739350
[235] 0.2353579 0.2818055 0.2473089 0.2793681 0.2859952 0.2755730 0.2600344 0.2577974 0.2313074 0.4615884 0.4600988 0.4591063 0.4198652 0.4244795 0.4281141 0.4226035 0.4206386 0.4292147
[253] 0.4267259 0.4337486 0.4216207 0.4262408 0.4303637 0.4274671 0.4218937 0.4263084 0.4178009 0.3916835 0.3986227 0.3930787 0.3977908 0.3927441 0.3921727 0.3919439 0.3908616 0.3915276
[271] 0.3845447 0.3958660 0.3893311 0.3925250 0.3875441 0.3927499 0.3804572 0.5016549 0.5012013 0.5090696 0.5060413 0.4782895 0.3979316 0.3691211 0.4231428 0.3672408 0.2841555 0.4057248
[289] 0.3888018 0.3659097 0.3887534 0.3623376 0.3988731 0.3968721 0.3948259 0.3590153 0.3988787 0.3609444 0.3524646 0.3603447 0.3595331 0.3585730 0.3509037 0.3533867 0.3692998 0.3678701
[307] 0.3713830 0.3673308 0.3699822 0.3672459 0.3604704 0.3470594 0.4936290 0.4846511 0.4937867 0.4517705 0.2310049 0.2087417 0.2236381 0.1961742 0.2095205 0.2133585 0.1852804 0.2182548
[325] 0.1926758 0.2618794 0.2205215 0.2255054 0.2366318 0.1912018 0.2118861 0.4679386 0.4635185 0.4411043 0.4234297 0.4201381 0.4259970 0.4128596 0.4168836 0.4239103 0.4188207 0.4178412
[343] 0.4110824 0.4194226 0.4205830 0.4149692 0.4251447 0.4255670 0.4267103 0.3611232 0.3562412 0.3712363 0.3442412 0.3668304 0.3530934 0.3678192 0.3678768 0.3515870 0.3597254 0.3603614
[361] 0.3413632 0.3419427 0.3676444 0.3581327 0.3448214 0.3504564 0.3566058 0.3487803 0.3563275 0.3526094 0.3450894 0.3524258 0.3419663 0.3558274 0.3509225 0.3524690 0.3497740 0.3659287
[379] 0.3460387 0.4186149 0.4315841 0.4196975 0.4142954 0.4272276 0.4220581 0.4237472 0.4237661 0.4246864 0.4101221 0.4252443 0.4297858 0.4326623 0.4259567 0.4291577 0.2363466 0.1910486
[397] 0.2086093 0.2369651 0.2038637 0.2544728 0.2414272 0.2393905 0.2225038 0.2322584 0.2416522 0.2338668 0.2226538 0.2313073 0.2465563 0.2219365 0.2574623 0.2559964 0.2548630 0.2434608
[415] 0.2871247 0.2602922 0.2597568 0.2582569 0.2696114 0.2577723 0.2594514 0.2572152 0.2509934 0.2176683 0.2275663 0.2251930 0.2266073 0.2271369 0.2169570 0.2411744 0.2112019 0.2195732
[433] 0.2221563 0.2232829 0.1962145 0.2118318 0.2516127 0.2430133 0.2631131 0.2298082 0.2515936 0.2392990 0.2432385 0.2488558 0.2573293 0.2709527 0.2426554 0.2671579 0.2662636 0.2748303
[451] 0.2224785 0.2580088 0.2470324 0.2634908 0.4650387 0.4685421 0.4281665 0.4091938 0.4089690 0.4045210 0.4082926 0.4060756 0.4081337 0.4070808 0.4094892 0.4030051 0.4036429 0.4134862
[469] 0.4034735 0.4154150 0.4083692 0.4148218 0.3539352 0.3598113 0.3540527 0.3573873 0.3649318 0.3451853 0.3355538 0.3580202 0.3532721 0.3573665 0.3641106 0.3563792 0.3523415 0.3563985
[487] 0.3574177 0.3701122 0.3523329 0.3646444 0.3612098 0.3583992 0.3555659 0.3709085 0.3518126 0.3516145 0.3414150 0.3529609 0.3661834 0.3551919 0.3525644 0.3640332 0.4071818 0.3967536
[505] 0.3799720 0.4161375 0.4142353 0.4020766 0.4111853 0.4010004 0.4031748 0.4107778 0.4011318 0.4057403 0.3997106 0.4079328 0.4144587 0.2605003 0.2797827 0.2599739 0.2659934 0.2595775
[523] 0.2811484 0.2789166 0.2714244 0.2377298 0.2646526 0.2668398 0.2620849 0.2688872 0.2634987 0.2582253 0.4082986 0.3943910 0.4309828 0.4016048 0.4046832 0.4689274 0.4829680 0.4506270
[541] 0.3172573 0.1955575 0.2735856 0.3523341 0.3769071 0.4503725 0.4765927 0.2692258 0.2595910 0.2754180 0.4301141 0.4878344 0.4779121 0.4524149 0.4761561 0.4372146 0.3746027 0.4318029
[559] 0.4324379 0.3444640 0.2238289 0.1923227 0.4754487 0.4731791 0.4765927 0.4356104 0.4483053 0.4089247
  [1] 0.2100 0.4400 0.3800 0.4100 0.3800 0.3700 0.3800 0.3700 0.3700 0.3700 0.3800 0.3500 0.3700 0.3700 0.3800 0.3600 0.3800 0.3800 0.3700 0.3400 0.3600 0.3400 0.3500 0.3500 0.3500 0.3300
 [27] 0.3400 0.3500 0.3300 0.3400 0.3600 0.3400 0.3500 0.3600 0.2800 0.2200 0.2700 0.2300 0.1700 0.2100 0.3100 0.1900 0.2400 0.2200 0.1900 0.2500 0.2100 0.2800 0.2400 0.3000 0.2200 0.1300
 [53] 0.3300 0.4900 0.4900 0.4800 0.4800 0.4100 0.4000 0.4000 0.4100 0.4000 0.4100 0.4000 0.4000 0.4000 0.4000 0.4100 0.4200 0.3900 0.4100 0.4100 0.4400 0.4500 0.4400 0.4300 0.4400 0.4500
 [79] 0.4400 0.4400 0.4500 0.4400 0.4400 0.4500 0.4400 0.4400 0.4500 0.4400 0.4500 0.4400 0.4500 0.4400 0.4500 0.4400 0.4400 0.4400 0.4400 0.4500 0.4500 0.4400 0.4500 0.4400 0.4100 0.4000
[105] 0.4000 0.4100 0.4100 0.4000 0.4200 0.3900 0.3900 0.4000 0.4100 0.4000 0.4200 0.4100 0.4000 0.4400 0.3600 0.4200 0.3200 0.3900 0.3900 0.4300 0.4400 0.3700 0.4400 0.4200 0.4200 0.4100
[131] 0.4200 0.4200 0.3000 0.2800 0.2000 0.3500 0.3900 0.3400 0.3400 0.3400 0.3500 0.3600 0.3500 0.3300 0.3600 0.3400 0.3600 0.3200 0.3600 0.3400 0.3700 0.3200 0.3300 0.3400 0.3500 0.3400
[157] 0.3500 0.3600 0.3200 0.3600 0.3200 0.3300 0.3300 0.3700 0.3500 0.3500 0.3200 0.4000 0.3200 0.3500 0.3000 0.4100 0.4600 0.3400 0.4700 0.4700 0.4700 0.3100 0.3900 0.3800 0.3600 0.3800
[183] 0.3700 0.3800 0.3700 0.3900 0.3900 0.3900 0.3700 0.3500 0.3800 0.3800 0.3800 0.5400 0.4500 0.4700 0.4200 0.4300 0.4200 0.4400 0.4300 0.4200 0.4300 0.4200 0.4200 0.4300 0.4300 0.4300
[209] 0.4100 0.4300 0.4300 0.3900 0.4000 0.3800 0.3700 0.3800 0.3900 0.3800 0.3800 0.3900 0.3900 0.3700 0.3800 0.3500 0.3800 0.4000 0.4900 0.4000 0.2400 0.2400 0.3200 0.3200 0.2500 0.2800
[235] 0.2200 0.3000 0.2400 0.2800 0.2900 0.2900 0.2600 0.2400 0.2100 0.4600 0.4600 0.4600 0.4300 0.4300 0.4300 0.4200 0.4300 0.4300 0.4300 0.4400 0.4200 0.4300 0.4300 0.4300 0.4300 0.4300
[261] 0.4300 0.3800 0.4000 0.3800 0.4000 0.3900 0.3900 0.3900 0.3900 0.3900 0.3800 0.4000 0.3900 0.3900 0.3800 0.3900 0.4200 0.5000 0.5100 0.5200 0.5100 0.4900 0.4300 0.3200 0.4400 0.4200
[287] 0.1800 0.4200 0.3700 0.3800 0.3700 0.3600 0.3800 0.4100 0.2900 0.4300 0.4000 0.3600 0.3500 0.3600 0.3600 0.3600 0.3400 0.3900 0.3800 0.3700 0.3800 0.3700 0.3800 0.3700 0.3600 0.3400
[313] 0.5000 0.5000 0.5000 0.4700 0.1900 0.1800 0.2300 0.1700 0.2100 0.2800 0.1600 0.2200 0.1700 0.2800 0.2300 0.1300 0.2600 0.1700 0.2100 0.4800 0.4600 0.4400 0.4300 0.4100 0.4300 0.4100
[339] 0.4100 0.4300 0.4200 0.4100 0.4200 0.4200 0.4200 0.4200 0.4300 0.4300 0.4300 0.3700 0.3800 0.3800 0.3600 0.3700 0.3500 0.3700 0.3700 0.3800 0.3800 0.3600 0.3300 0.3300 0.3800 0.3600
[365] 0.3400 0.3500 0.3600 0.3500 0.3500 0.3500 0.3400 0.3700 0.3300 0.3600 0.3600 0.3600 0.3500 0.3700 0.3500 0.4300 0.4400 0.4000 0.4100 0.4300 0.4100 0.4200 0.4000 0.4300 0.4000 0.4000
[391] 0.4300 0.4400 0.4100 0.4300 0.2200 0.1600 0.1900 0.2400 0.1900 0.2700 0.2500 0.2500 0.2100 0.2700 0.2400 0.2000 0.2200 0.2400 0.2600 0.1900 0.2700 0.2600 0.2400 0.2400 0.3100 0.3100
[417] 0.1600 0.2600 0.2800 0.2600 0.2700 0.2600 0.2600 0.1900 0.2300 0.2200 0.2800 0.2600 0.2000 0.2500 0.2000 0.2000 0.2000 0.2300 0.1700 0.1900 0.2700 0.2500 0.2900 0.2200 0.2500 0.2300
[443] 0.2400 0.2500 0.2600 0.2700 0.2400 0.2800 0.3000 0.2900 0.2000 0.2800 0.2200 0.2800 0.4700 0.4900 0.4400 0.4100 0.4100 0.4000 0.4000 0.4100 0.4100 0.4000 0.4100 0.4000 0.4000 0.4200
[469] 0.4000 0.4200 0.4000 0.4200 0.3500 0.3600 0.3500 0.3600 0.3700 0.3400 0.3200 0.3600 0.3500 0.3600 0.3700 0.3500 0.3500 0.3600 0.3700 0.3800 0.3500 0.3700 0.3600 0.3600 0.3500 0.3800
[495] 0.3500 0.3500 0.3300 0.3500 0.3700 0.3800 0.3500 0.3700 0.4000 0.3900 0.3600 0.4200 0.4200 0.4000 0.4200 0.4000 0.4100 0.3900 0.4000 0.4000 0.4000 0.4100 0.4200 0.3000 0.2900 0.1500
[521] 0.3100 0.2600 0.2900 0.3000 0.2800 0.2200 0.2700 0.2700 0.2100 0.3100 0.2200 0.2900 0.4249 0.4247 0.4438 0.4309 0.4163 0.5036 0.5266 0.4806 0.2993 0.1456 0.2758 0.3623 0.3768 0.4900
[547] 0.4800 0.1926 0.2355 0.1979 0.4837 0.5045 0.4868 0.4546 0.4965 0.4148 0.4023 0.4305 0.3370 0.3539 0.2480 0.1717 0.5105 0.4794 0.4800 0.4548 0.4485 0.4037

model_per

Random Forest 

397 samples
106 predictors

No pre-processing
Resampling: Cross-Validated (10 fold) 
Summary of sample sizes: 357, 358, 357, 355, 357, 359, ... 
Resampling results across tuning parameters:

  mtry  splitrule   RMSE        Rsquared   MAE       
    2   variance    0.03663598  0.8118797  0.02423285
    2   extratrees  0.03786082  0.8096785  0.02568269
   54   variance    0.03310826  0.8322125  0.02154125
   54   extratrees  0.03269921  0.8385388  0.02138579
  106   variance    0.03410869  0.8203200  0.02174119
  106   extratrees  0.03246248  0.8399941  0.02118672

Tuning parameter 'min.node.size' was held constant at a value of 5
RMSE was used to select the optimal model using the smallest value.
The final values used for the model were mtry = 106, splitrule = extratrees and min.node.size = 5.

# install.packages("gridExtra")
library(gridExtra)
library(parallel)

plot_column <- function(column, importance){
  results <- data.frame(x = dataset$tmg, y = dataset[[column]])
  plt <- ggplot(results, aes(x = x, y = y)) +
    geom_point(color='blue') +
     xlab("") +
    ylab(column) +
    theme_bw()
  return(plt)  
}

plot_important <- function(important, title, n_to_plot=6, columns=2) grid.arrange(top=title, grobs = lapply(important$predictor[1:n_to_plot], plot_column, importance=important$importance[1:n_to_plot]), ncol = columns)

# plot_list1 <- lapply(top1$predictor[1:6], plot_column)
# grid.arrange(top="Alpha=1", grobs = plot_list1, ncol = 2)

# plot_list0 <- lapply(top0$predictor[1:6], plot_column)
# grid.arrange(top="Alpha=0", grobs = plot_list0, ncol = 2)
top0 <- model_imp
png("rf_permutation_improtance.png", width = 1000, height = 1200)
plot_important(top0, n_to_plot=20, columns=4, "RF Permutation Importance")
dev.off()

null device 
          1 

# Install and load necessary libraries
# install.packages("ggplot2")
library(ggplot2)

# Your data
data <- read.table(text = "  mtry  splitrule   RMSE        Rsquared   MAE       
    2   variance    0.03663598  0.8118797  0.02423285
    2   extratrees  0.03786082  0.8096785  0.02568269
   54   variance    0.03310826  0.8322125  0.02154125
   54   extratrees  0.03269921  0.8385388  0.02138579
  106   variance    0.03410869  0.8203200  0.02174119
  106   extratrees  0.03246248  0.8399941  0.02118672", header = TRUE)

# Plotting using ggplot2
ggplot(data, aes(x = factor(mtry), y = RMSE, fill = splitrule)) +
  geom_bar(stat = "identity", position = "dodge") +
  labs(title = "Random Forest RMSE by mtry and splitrule",
       x = "mtry",
       y = "RMSE") +
  theme_minimal()